Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.
Open accessPublished version of a preprint
Abstract
We show that faces contain much more information about sexual orientation than can be perceived or interpreted by the human brain. We used deep neural networks to extract features from 35,326 facial images. These features were entered into a logistic regression aimed at classifying sexual orientation. Given a single facial image, a classifier could correctly distinguish between gay and heterosexual men in 81% of cases, and in 71% of cases for women. Human judges achieved much lower accuracy: 61% for men and 54% for women. The accuracy of the algorithm increased to 91% and 83%, respectively, given five facial images per person. Facial features employed by the classifier included both fixed (e.g., nose shape) and transient facial features (e.g., grooming style). Consistent with the prenatal hormone theory of sexual orientation, gay men and women tended to have gender-atypical facial morphology, expression, and grooming styles. Prediction models aimed at gender alone allowed for detecting gay males with 57% accuracy and gay females with 58% accuracy. Those findings advance our understanding of the origins of sexual orientation and the limits of human perception. Additionally, given that companies and governments are increasingly using computer vision algorithms to detect people's intimate traits, our findings expose a threat to the privacy and safety of gay men and women. (PsycINFO Database Record
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
A recent paper used publicly available images of faces uploaded to a dating website to train AI software to detect someone's sexuality based on a small number of photos, raising questions around consent.
“For example, a recent paper used publicly available images of faces uploaded to a dating website as a way to train AI software to detect someone’s sexuality based on a small number of photos (Wang and Kosinski 2018). While the study received ethics committee approval, it raises further questions around consent, since it is implausible that the users of the dating website could or necessarily would”
From How to Design AI for Social Good: Seven Essential Factors (2020) · cites this paper as Wang and Kosinski 2018
Computer vision algorithms have been used to ascertain sexual orientation of people by scrutinizing thousands of facial images taken from public profiles on a dating website.
“For example, a study found that computer vision algorithms were used to ascertain sexual orientation of people by scrutinizing thousands of facial images taken from public profiles on a dating website (Wang and Kosinski, 2017).”
From SHIFTing artificial intelligence to be responsible in healthcare: a systematic review (2022) · cites this paper as Wang and Kosinski, 2017
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Hao, X., Zhang, G., & Ma, S. (2016) · International Journal of Semantic Computing · doi:10.1142/s1793351x16500045 · Cited by 82,372
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362 · see citing works
Parkhi, O. M., Vedaldi, A., & Zisserman, A. (2015) · Procedings of the British Machine Vision Conference 2015 · doi:10.5244/c.29.41 · Cited by 5,479 · see citing works
Private traits and attributes are predictable from digital records of human behavior
Kosinski, M., Stillwell, D., & Graepel, T. (2013) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1218772110 · Cited by 2,829 · see citing works
Self-citationThe Evolutionary Psychology of Facial Beauty
Rhodes, G. (2005) · Annual Review of Psychology · doi:10.1146/annurev.psych.57.102904.190208 · Cited by 1,757 · see citing works
Inferences of competence from faces predict election outcomes
Todorov, A., Mandisodza, A. N., Goren, A., & Hall, C. C. (2005) · Science · doi:10.1126/science.1110589 · Cited by 1,508 · see citing works
Social Cognition: Thinking Categorically about Others
Macrae, C. N., & Bodenhausen, G. V. (2000) · Annual Review of Psychology · doi:10.1146/annurev.psych.51.1.93 · Cited by 1,347 · see citing works
The evolution of human skin coloration
Jablonski, N. G., & Chaplin, G. (2000) · Journal of Human Evolution · doi:10.1006/jhev.2000.0403 · Cited by 1,212 · see citing works
Kosiński, M., Matz, S., Gosling, S. D., Popov, V., & Stillwell, D. J. (2015) · American Psychologist · doi:10.1037/a0039210 · Cited by 926 · see citing works
Self-citationUnderstanding evaluation of faces on social dimensions
Todorov, A., Said, C. P., Engell, A. D., & Oosterhof, N. N. (2008) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2008.10.001 · Cited by 706 · see citing works
Predicting political elections from rapid and unreflective face judgments
Ballew, C. C., & Todorov, A. T. (2007) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0705435104 · Cited by 473 · see citing works
Facial appearance affects voting decisions.
Little, A., Burriss, R., Jones, B., & Roberts, S. C. (2007) · Evolution and Human Behavior · doi:10.1016/j.evolhumbehav.2006.09.002 · Cited by 420 · see citing works
Accuracy of judgments of sexual orientation from thin slices of behavior.
Ambady, N., Hallahan, M., & Conner, B. (1999) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.77.3.538 · Cited by 380
Sexual orientation and the size of the anterior commissure in the human brain.
Allen, L. S., & Gorski, R. A. (1992) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.89.15.7199 · Cited by 367 · see citing works
Sex-related variation in human behavior and the brain
Hines, M. (2010) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2010.07.005 · Cited by 355 · see citing works
Accurate Social Perception at Zero Acquaintance: The Affordances of a Gibsonian Approach
Zebrowitz, L. A., & Collins, M. A. (1997) · Personality and Social Psychology Review · doi:10.1207/s15327957pspr0103_2 · Cited by 336 · see citing works
Brief exposures: Male sexual orientation is accurately perceived at 50ms
Rule, N. O., & Ambady, N. (2007) · Journal of Experimental Social Psychology · doi:10.1016/j.jesp.2007.12.001 · Cited by 332 · see citing works
Brown, E., & Perrett, D. I. (1993) · Perception · doi:10.1068/p220829 · Cited by 330 · see citing works
Telling facial metrics: facial width is associated with testosterone levels in men
Lefevre, C. E., Lewis, G. J., Perrett, D. I., & Penke, L. (2013) · Evolution and Human Behavior · doi:10.1016/j.evolhumbehav.2013.03.005 · Cited by 295 · see citing works
Penton‐Voak, I. S., Pound, N., Little, A. C., & Perrett, D. I. (2006) · Social Cognition · doi:10.1521/soco.2006.24.5.607 · Cited by 290 · see citing works
Biological Limits of Gender Construction
Udry, J. (2000) · American Sociological Review · doi:10.1177/000312240006500307 · Cited by 254
Sell, R. L., Wells, J. A., & Wypij, D. (1995) · Archives of Sexual Behavior · doi:10.1007/bf01541598 · Cited by 234 · see citing works
Ontogeny of facial dimorphism and patterns of individual development within one human population
Булыгина, Е. С., Mitterœcker, P., & Aiello, L. C. (2006) · American Journal of Physical Anthropology · doi:10.1002/ajpa.20317 · Cited by 232 · see citing works
Accuracy and awareness in the perception and categorization of male sexual orientation.
Rule, N. O., Ambady, N., Adams, R. B., & Macrae, C. N. (2008) · Journal of Personality and Social Psychology · doi:10.1037/a0013194 · Cited by 223 · see citing works
Gaydar: Facebook Friendships Expose Sexual Orientation
Jernigan, C., & Mistree, B. F. T. (2009) · First Monday · doi:10.5210/fm.v14i10.2611 · Cited by 222 · see citing works
Extraversion is accurately perceived after a 50-ms exposure to a face
Borkenau, P., Brecke, S., Möttig, C., & Paelecke, M. (2009) · Journal of Research in Personality · doi:10.1016/j.jrp.2009.03.007 · Cited by 208 · see citing works
Prenatal testosterone exposure is related to sexually dimorphic facial morphology in adulthood
Whitehouse, A., Gilani, S. Z., Faisal Shafait, B., Mian, A. S., Tan, D. W., Maybery, M. T., Keelan, J. A., Hart, R. J., Handelsman, D. J., Goonawardene, M., et al. (2015) · Proceedings of the Royal Society B Biological Sciences · doi:10.1098/rspb.2015.1351 · Cited by 181 · see citing works
Astley, S. J., Stachowiak, J. A., Clarren, S. K., & Clausen, C. L. (2002) · The Journal of Pediatrics · doi:10.1067/mpd.2002.129030 · Cited by 163 · see citing works
Diagnostically relevant facial gestalt information from ordinary photos
Ferry, Q. R. V., Steinberg, J., Webber, C., FitzPatrick, D. R., Ponting, C. P., Zisserman, A., & Nellåker, C. (2014) · eLife · doi:10.7554/elife.02020 · Cited by 159 · see citing works
Psychoendocrine research on sexual orientation. Current status and future options.
Meyer-Bahlburg, H. (1984) · Progress in Brain Research · doi:10.1016/s0079-6123(08)64448-9 · Cited by 153 · see citing works
Democrats and Republicans Can Be Differentiated from Their Faces
Rule, N. O., & Ambady, N. (2010) · PLoS ONE · doi:10.1371/journal.pone.0008733 · Cited by 149 · see citing works
Accuracy in social perception: contributions of facial and vocal information.
Berry, D. S. (1991) · Journal of Personality and Social Psychology · doi:10.1037//0022-3514.61.2.298 · Cited by 134 · see citing works
Feature Saliency in Judging the Sex and Familiarity of Faces
Roberts, T., & Bruce, V. (1988) · Perception · doi:10.1068/p170475 · Cited by 132 · see citing works
Political Ideology at Face Value
Samochowiec, J., Wänke, M., & Fiedler, K. (2010) · Social Psychological and Personality Science · doi:10.1177/1948550610372145 · Cited by 90 · see citing works
The Kernel of Truth in Judgments of Deceptiveness
Bond, C. F., Berry, D. S., & Atoum, A. (1994) · Basic and Applied Social Psychology · doi:10.1207/s15324834basp1504_8 · Cited by 78 · see citing works
Ambiguous Group Membership Is Extracted Automatically From Faces
Rule, N. O., Macrae, C., & Ambady, N. (2009) · Psychology Science · doi:10.1111/j.1467-9280.2009.02314.x · Cited by 77
Facial Width-To-Height Ratio Relates to Alpha Status and Assertive Personality in Capuchin Monkeys
Lefevre, C. E., Wilson, V. A. D., Morton, F. B., Brosnan, S. F., Paukner, A., & Bates, T. C. (2014) · PLoS ONE · doi:10.1371/journal.pone.0093369 · Cited by 67 · see citing works
Male Homosexuality: Nature or Culture?
Jannini, E. A., Blanchard, R., Camperio-Ciani, A., & Bancroft, J. (2010) · The Journal of Sexual Medicine · doi:10.1111/j.1743-6109.2010.02024.x · Cited by 62 · see citing works
Facial Structure Predicts Sexual Orientation in Both Men and Women
Skorska, M. N., Geniole, S., Vrysen, B. M., et al. (2015) · Archives of Sexual Behavior · doi:10.1007/s10508-014-0454-4 · Cited by 58
Shape Differences Between the Faces of Homosexual and Heterosexual Men
Valentová, J. V., Kleisner, K., Havlı́ček, J., & Neustupa, J. (2013) · Archives of Sexual Behavior · doi:10.1007/s10508-013-0194-x · Cited by 56 · see citing works
Facial coloration tracks changes in women's estradiol
Jones, B. C., Hahn, A. C., Fisher, C. I., Wincenciak, J., Kandrik, M., Roberts, S. C., Little, A. C., & DeBruine, L. M. (2015) · Psychoneuroendocrinology · doi:10.1016/j.psyneuen.2015.02.021 · Cited by 54 · see citing works
Detection of Sexual Orientation (“Gaydar”) by Homosexual and Heterosexual Women
Lyons, M. T., Lynch, A., Brewer, G., & Bruno, D. (2013) · Archives of Sexual Behavior · doi:10.1007/s10508-013-0144-7 · Cited by 50 · see citing works
Gender and gender role differences in smiling and communication consistency
Halberstadt, A., Hayes, C. W., & Pike, K. M. (1988) · Sex Roles · doi:10.1007/bf00289738 · Cited by 49 · see citing works
Emotions facilitate the communication of ambiguous group memberships.
Tskhay, K. O., & Rule, N. O. (2015) · Emotion · doi:10.1037/emo0000077 · Cited by 49 · see citing works
Pythagoras: His Life, Teaching, and Influence
Servatius, B. (2009) · doi:10.4169/193113409x469767 · Cited by 41
Familiarity increases the accuracy of categorizing male sexual orientation
Brambilla, M., Riva, P., & Rule, N. O. (2013) · doi:10.1016/j.paid.2013.02.023 · Cited by 36
Hughes, S. M., & Bremme, R. (2011) · Journal of Social Evolutionary and Cultural Psychology · doi:10.1037/h0099261 · Cited by 31 · see citing works
Masculinity-femininity predicts sexual orientation in men but not in women.
Udry, J., & Chantala, K. (2006) · Journal of Biosocial Science · doi:10.1017/s002193200500101x · Cited by 30
Tomboy Chic: Re-Fashioning Gender Rebellion
Skerski, J. (2011) · Journal of Lesbian Studies · doi:10.1080/10894160.2011.532031 · Cited by 14
Dress for Success: Human Facial Expressions are Important Signals of Emotions
Lõhmus, M., Sundström, L., & Björklund, M. (2009) · Annales Zoologici Fennici · doi:10.5735/086.046.0108 · Cited by 11 · see citing works
Glimcher, M. E., García, R. I., & Szabó, G. (1978) · Journal of Experimental Zoology · doi:10.1002/jez.1402040210 · Cited by 10 · see citing works
Flagged first, then by citations
14 of the 66 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
How artificial intelligence will change the future of marketing
Davenport, T., Guha, A., Grewal, D., & Bressgott, T. (2020) · Journal of the Academy of Marketing Science · doi:10.1007/s11747-019-00696-0 · Cited by 2,875 · see citing works
“For example, two Stanford researchers used deep neural networks to identify people ’ s sexual orientation, merely by analyzing facial images (Wang and Kosinski 2018).”
BackgroundThe Ethics of AI Ethics: An Evaluation of Guidelines
Hagendorff, T. (2020) · Minds and Machines · doi:10.1007/s11023-020-09517-8 · Cited by 1,911 · see citing works
Taxonomy of Risks posed by Language Models
Weidinger, L., Uesato, J., Rauh, M., Griffin, C., Huang, P.-S., Mellor, J., Glaese, A., Cheng, M., Balle, B., Kasirzadeh, A., et al. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533088 · Cited by 764 · see citing works
“Such systems may never be accurate - for example, image classification models that attempt to infer unobservable characteristics, such as sexual orientation from a portrait [190], are inherently prone to error.”
BackgroundTowards a standard for identifying and managing bias in artificial intelligence
Schwartz, R., Vassilev, A., Greene, K., Perine, L., Burt, A., & Hall, P. (2022) · doi:10.6028/nist.sp.1270 · Cited by 548 · see citing works
The Ethics of Artificial Intelligence
Floridi, L. (2023) · doi:10.1093/oso/9780198883098.001.0001 · Cited by 608 · see citing works
An Overview of Artificial Intelligence Ethics
Huang, C., Zhang, Z., Mao, B., & Yao, X. (2022) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2022.3194503 · Cited by 583 · see citing works
“of AI: AI technologies, such as facial recognition and image generation, can work better than humans [34].”
BackgroundChun, W. H. K., & Barnett, A. (2021) · The MIT Press eBooks · doi:10.7551/mitpress/14050.001.0001 · Cited by 438 · see citing works
How to Design AI for Social Good: Seven Essential Factors
Floridi, L., Cowls, J., King, T. C., & Taddeo, M. (2020) · Science and Engineering Ethics · doi:10.1007/s11948-020-00213-5 · Cited by 430 · see citing works
“For example, a recent paper used publicly available images of faces uploaded to a dating website as a way to train AI software to detect someone’s sexuality based on a small number of photos (Wang and Kosinski 2018).”
Assessing gender bias in machine translation: a case study with Google Translate
Prates, M. O. R., Avelar, P. H., & Lamb, L. C. (2020) · Neural Computing and Applications · doi:10.1007/s00521-019-04144-6 · Cited by 410 · see citing works
“ectly revived heated debates about the controversial eld of physiognomy, with proposals of AI systems capable of identifying the sexual orientation of an individual through its facial characteristics [38]. Similar concerns are growing at an unprecedented rate in the media, with reports of Apple’s Iphone X face unlock…”
BackgroundLarge-scale GWAS reveals insights into the genetic architecture of same-sex sexual behavior
Ganna, A., Verweij, K. J. H., Nivard, M. G., Maier, R., Wedow, R., Busch, A. S., Abdellaoui, A., Guo, S., Sathirapongsasuti, J. F., 23andMe Research Team, et al. (2019) · Science · doi:10.1126/science.aat7693 · Cited by 406 · see citing works
The Ethics of AI Ethics: An Evaluation of Guidelines
Hagendorff, T. (2020) · PhilPapers (PhilPapers Foundation) · Cited by 394 · see citing works
Misinformation in and about science
West, J. D., & Bergstrom, C. T. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1912444117 · Cited by 391 · see citing works
“Controversial work on detecting sexual orientation from photographs using artificial intelligence (32) was reported in The Economist and The Guardian before a preprint or white paper was available.”
BackgroundSHIFTing artificial intelligence to be responsible in healthcare: a systematic review
Siala, H., & Wang, Y. (2022) · Social Science & Medicine · doi:10.1016/j.socscimed.2022.114782 · Cited by 358 · see citing works
Self-citationHow artificial intelligence will affect the future of retailing
Guha, A., Grewal, D., Kopalle, P. K., Haenlein, M., Schneider, M. J., Jung, H., Moustafa, R., Hegde, D. R., & Hawkins, G. (2021) · Journal of Retailing · doi:10.1016/j.jretai.2021.01.005 · Cited by 365 · see citing works
Kopalle, P. K., Gangwar, M., Kaplan, A., Ramachandran, D., Reinartz, W., & Rindfleisch, A. (2021) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2021.11.002 · Cited by 323 · see citing works
A critical review of algorithms in HRM: Definition, theory, and practice
Cheng, M. M., & Hackett, R. D. (2021) · Human Resource Management Review · doi:10.1016/j.hrmr.2019.100698 · Cited by 283 · see citing works
Chou, Y.-L., Moreira, C., Bruza, P., Ouyang, C., & Jorge, J. (2022) · Information Fusion · doi:10.1016/j.inffus.2021.11.003 · Cited by 302 · see citing works
“Other studies revealed gender and racial biases in automated facial analysis algorithms made available by commercial companies [11], gender biases in textual predictive models [14, 15, 16] or even more discriminatory topics such as facial features according to sexual orientation [17].”
BackgroundArtificial Intelligence Crime: An Interdisciplinary Analysis of Foreseeable Threats and Solutions
King, T. C., Aggarwal, N., Taddeo, M., & Floridi, L. (2020) · Science and Engineering Ethics · doi:10.1007/s11948-018-00081-0 · Cited by 271 · see citing works
EPM 20/20: A Review, Framework, and Research Agenda for Electronic Performance Monitoring
Ravid, D. M., Tomczak, D. L., White, J. C., & Behrend, T. S. (2020) · Journal of Management · doi:10.1177/0149206319869435 · Cited by 277 · see citing works
“Wang and Kosinski (2018) recently showed that facial detection software could be used to identify an individual’s sexual orientation with relative accuracy, Yaden and colleagues (2018) used linguistic patterns to predict religious affiliation, and online behaviors have been shown to reliably…”
MethodYang, K., Qinami, K., Fei-Fei, L., Deng, J., & Russakovsky, O. (2020) · Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency · doi:10.1145/3351095.3375709 · Cited by 254 · see citing works
“Examples range from quantifying beauty [47] to predicting sexual orientation [76] to (arguably) annotating gender [25, 42].”
BackgroundConsiderations for ethics review of big data health research: A scoping review
Ienca, M., Ferretti, A., Hurst, S., Puhan, M. A., Lovis, C., & Vayena, E. (2018) · PLoS ONE · doi:10.1371/journal.pone.0204937 · Cited by 263 · see citing works
Understanding from Machine Learning Models
Sullivan, E. (2019) · The British Journal for the Philosophy of Science · doi:10.1093/bjps/axz035 · Cited by 229 · see citing works
“Researchers developed a facial recognition model that seeks to identify the sexual orientation of individuals (Wang and Kosinski [2018]).”
BackgroundConsumer privacy and the future of data-based innovation and marketing
Bleier, A., Goldfarb, A., & Tucker, C. (2020) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2020.03.006 · Cited by 239 · see citing works
“…Acquisti and Gross (2009) showed that public data can predict social security numbers, Crandall et al. (2010) showed that hidden social connections can be inferred from public social media, and Wang and Kosinski (2018) claimed to be able to identify sexual orientation from Facebook profile pictures.”
Social Robots on a Global Stage: Establishing a Role for Culture During Human–Robot Interaction
Lim, V., Rooksby, M., & Cross, E. S. (2020) · International Journal of Social Robotics · doi:10.1007/s12369-020-00710-4 · Cited by 228 · see citing works
Re-imagining Algorithmic Fairness in India and Beyond
Sambasivan, N., Arnesen, E., Hutchinson, B., Doshi, T., & Prabhakaran, V. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445896 · Cited by 228 · see citing works
“, [187, 212]) potentially being exported to India and harming minorities.”
BackgroundFacial recognition technology can expose political orientation from naturalistic facial images
Kosiński, M. (2021) · Scientific Reports · doi:10.1038/s41598-020-79310-1 · Cited by 205 · see citing works
Self-citationAlmeida, D., Shmarko, K., & Lomas, E. (2022) · AI and Ethics · doi:10.1007/s43681-021-00077-w · Cited by 203 · see citing works
“The extent to which they can be relied on without human checks is one of constant concern, particularly as the use of these technologies as well as identifying individuals is extending their reach to make further judgements about individuals including in regard to their behaviours, motivations, emotions, and protected…”
BackgroundINTERNATIONAL HUMAN RIGHTS LAW AS A FRAMEWORK FOR ALGORITHMIC ACCOUNTABILITY
McGregor, L., Murray, D., & Ng, V. (2019) · International and Comparative Law Quarterly · doi:10.1017/s0020589319000046 · Cited by 207 · see citing works
Designing for human rights in AI
Aizenberg, E., & van den Hoven, J. (2020) · Big Data & Society · doi:10.1177/2053951720949566 · Cited by 220 · see citing works
The effects of trait-based personalization in social media advertising
Winter, S., Masłowska, E., & Vos, A. L. (2020) · Computers in Human Behavior · doi:10.1016/j.chb.2020.106525 · Cited by 171 · see citing works
Das, A., Dantcheva, A., & Brémond, F. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11009-3_35 · Cited by 177 · see citing works
“Moreover nowadays, FA is not restricted to face recognition, but has transgressed onto emotion analysis [5], gender classification [6], as well as other facial characteristics [7–9].”
BackgroundAI Chatbots in Digital Mental Health
Balcombe, L. (2023) · Informatics · doi:10.3390/informatics10040082 · Cited by 170 · see citing works
How to Design AI for Social Good: Seven Essential Factors
Floridi, L., Cowls, J., King, T. C., & Taddeo, M. (2021) · Philosophical studies series · doi:10.1007/978-3-030-81907-1_9 · Cited by 171 · see citing works
Personality Research and Assessment in the Era of Machine Learning
Stachl, C., Pargent, F., Hilbert, S., Harari, G. M., Schoedel, R., Vaid, S. S., Gosling, S. D., & Bühner, M. (2020) · European Journal of Personality · doi:10.1002/per.2257 · Cited by 165 · see citing works
Ethical machines: The human-centric use of artificial intelligence
Lepri, B., Oliver, N., & Pentland, A. (2021) · iScience · doi:10.1016/j.isci.2021.102249 · Cited by 149 · see citing works
Put to the test: For a new sociology of testing
Marres, N., & Stark, D. (2020) · British Journal of Sociology · doi:10.1111/1468-4446.12746 · Cited by 149 · see citing works
Towards trustworthy LLMs: a review on debiasing and dehallucinating in large language models
Lin, Z., Guan, S., Zhang, W., Zhang, H., Li, Y., & Zhang, H. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10896-y · Cited by 141 · see citing works
Regulating Artificial Intelligence: Proposal for a Global Solution
Erdélyi, O. J., & Goldsmith, J. (2018) · Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3278721.3278731 · Cited by 133 · see citing works
“creasing privacy threat posed by pattern recognition applications (Kosinski and Wang 2017).”
BackgroundInterdisciplinary Research in Artificial Intelligence: Challenges and Opportunities
Kusters, R., Misevic, D., Berry, H., Cully, A., Cunff, Y. L., Dandoy, L., Díaz-Rodríguez, N., Ficher, M., Grizou, J., Othmani, A., et al. (2020) · Frontiers in Big Data · doi:10.3389/fdata.2020.577974 · Cited by 132 · see citing works
“…date, the technology is not ready [e.g., IBM ’ s Watson for oncology (Strickland, 2019)], and to ethically questionable applications [e.g., predicting sexual orientations from people ’ s faces (Wang and Kosinski, 2017), using facial recognition in law enforcement or for commercial use (Clearview)].”
BackgroundCooper, A. F., Moss, E., Laufer, B., & Nissenbaum, H. F. (2022) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3531146.3533150 · Cited by 144 · see citing works
“[60], a face dataset that has been created with utmost attention to privacy, may nevertheless cause privacy harm when a model derived from it is used to identify individuals as gay [141].”
BackgroundPrivacy in the age of psychological targeting
Matz, S., Appel, R. E., & Kosiński, M. (2019) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2019.08.010 · Cited by 127 · see citing works
Self-citationFairness for Unobserved Characteristics: Insights from Technological Impacts on Queer Communities
Tomasev, N., McKee, K. R., Kay, J., & Mohamed, S. (2021) · Proceedings of the AAAI/ACM Conference on AI Ethics and Society · doi:10.1145/3461702.3462540 · Cited by 139 · see citing works
“In 2017, Stanford researchers attempted to build an AI ‘gaydar’, a computer vision model capable of guessing a person’s sexual orientation from images (Wang and Kosinski 2018).”
BackgroundJin, B., Cruz, L., & Gonçalves, N. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3197235 · Cited by 107 · see citing works
“Because of its noninvasiveness, face recognition has become the most user-friendly biometric method, which leads to its wide applications [2], [3], [4].”
BackgroundLuoto, S., Krams, I. A., & Rantala, M. J. (2018) · Archives of Sexual Behavior · doi:10.1007/s10508-018-1261-0 · Cited by 103 · see citing works
Deep learning in electron microscopy
Ede, J. M. (2020) · Machine Learning Science and Technology · doi:10.1088/2632-2153/abd614 · Cited by 105 · see citing works
Digital hyperconnectivity and the self
Brubaker, R. (2020) · Theory and Society · doi:10.1007/s11186-020-09405-1 · Cited by 103 · see citing works
Hagendorff, T. (2021) · AI and Ethics · doi:10.1007/s43681-021-00122-8 · Cited by 117 · see citing works
Stark, L., & Crawford, K. (2019) · Surveillance & Society · doi:10.24908/ss.v17i3/4.10821 · Cited by 102 · see citing works
Assessing the Big Five personality traits using real-life static facial images
Kachur, A., Osin, E., Davydov, D., Shutilov, K., & Novokshonov, A. (2020) · Scientific Reports · doi:10.1038/s41598-020-65358-6 · Cited by 97 · see citing works
Privacy-Enhancing Technology and Everyday Augmented Reality
O’Hagan, J., Saeghe, P., Gugenheimer, J., Medeiros, D., Marky, K., Khamis, M., & McGill, M. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3569501 · Cited by 103 · see citing works
What demographic attributes do our digital footprints reveal? A systematic review
Hinds, J. M., & Joinson, A. (2018) · PLoS ONE · doi:10.1371/journal.pone.0207112 · Cited by 98 · see citing works
Privacy and Artificial Intelligence
Curzon, J., Kosa, T. A., Akalu, R., & El‐Khatib, K. (2021) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2021.3088084 · Cited by 102 · see citing works
Ethics review of big data research: What should stay and what should be reformed?
Ferretti, A., Ienca, M., Sheehan, M., Blasimme, A., Dove, E. S., Farsides, B., Friesen, P., Kahn, J., Karlen, W., Kleist, P., et al. (2021) · BMC Medical Ethics · doi:10.1186/s12910-021-00616-4 · Cited by 101 · see citing works
“As an example, computational technologies have shown potential to determine individuals’ sexual orientation by screening facial images [56].”
BackgroundTait, E., & Pierson, C. M. (2022) · Journal of the Australian Library and Information Association · doi:10.1080/24750158.2022.2081111 · Cited by 90 · see citing works
“To provide relevant information services into the future, issues of algorithmic in fl uence on human behaviour, bias and questions of ethical machine learning and AI are becoming ever more pressing and realistic (Ober-meyer et al., 2019; United Nations, 2021; Wang & Kosinski, 2018).”
BackgroundRegulating artificial intelligence: Proposal for a global solution
Erdélyi, O. J., & Goldsmith, J. (2022) · Government Information Quarterly · doi:10.1016/j.giq.2022.101748 · Cited by 93 · see citing works
Supervised machine learning methods in psychology: A practical introduction with annotated R code
Rosenbusch, H., Soldner, F., Evans, A. M., & Zeelenberg, M. (2021) · Social and Personality Psychology Compass · doi:10.1111/spc3.12579 · Cited by 86 · see citing works
Designing AI for Social Good: Seven Essential Factors
Cowls, J., King, T., Taddeo, M., & Floridi, L. (2019) · SSRN Electronic Journal · doi:10.2139/ssrn.3388669 · Cited by 83 · see citing works
“For example, a recent paper used publicly available images of faces uploaded to a dating website as a way to train AI software to detect someone’s sexuality based on a small number of photos (Wang and Kosinski 2018).”
BackgroundAnonymizing k Facial Attributes via Adversarial Perturbations
Chhabra, S., Singh, R., Mayank Vatsa, A., & Gupta, G. (2018) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2018/91 · Cited by 78 · see citing works
“In a recent research, Wang and Kosinksi predicted the “secrets of the heart”, such as predicting sexual orientation from face images [Wang and Kosinski, 2017].”
Telkamp, J. B., & Anderson, M. H. (2022) · Journal of Business Ethics · doi:10.1007/s10551-022-05057-6 · Cited by 73 · see citing works
Burr, C., & Cristianini, N. (2019) · Minds and Machines · doi:10.1007/s11023-019-09497-4 · Cited by 75 · see citing works
The Cambridge Companion to Queer Studies
Erin, G., Somerville, S. B., Amin, K., et al. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108699396 · Cited by 75 · see citing works
Kalogerakou, T., & Αντωνιάδου, Μ. (2024) · Antioxidants · doi:10.3390/antiox13121508 · Cited by 98 · see citing works
Deepfakes, Phrenology, Surveillance, and More! A Taxonomy of AI Privacy Risks
Lee, H.-P. (., Yang, Y.-J., Von Davier, T. S., Forlizzi, J., & Das, S. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642116 · Cited by 82 · see citing works
“…improve, we see new uses of these technologies to make spurious predictions about individuals and their behavior, portend-ing a new age of AI-facilitated phrenology and physiognomy: e.g., through the use of profle images to predict things like sexual orientation [145] and “criminality” [154].”
BackgroundFacial Recognition Neural Networks Confirm Success of Facial Feminization Surgery
Chen, K., Lu, S. M., Cheng, R., et al. (2019) · Plastic & Reconstructive Surgery · doi:10.1097/prs.0000000000006342 · Cited by 69 · see citing works
AI and recruiting software: Ethical and legal implications
Fernández-Martínez, C., & Fernández, A. (2020) · Paladyn J. Behav. Robotics · doi:10.1515/pjbr-2020-0030 · Cited by 66 · see citing works
Privacy Intelligence: A Survey on Image Privacy in Online Social Networks
Liu, C., Zhu, T., Zhang, J., & Zhou, W. (2022) · ACM Computing Surveys · doi:10.1145/3547299 · Cited by 84 · see citing works
“Meanwhile, the implicit information in images, such as health condition [11] and sex orientation [12], can be accurately captured by machine learning models.”
Situated algorithms: a sociotechnical systemic approach to bias
Draude, C., Klumbytė, G., Lücking, P., & Treusch, P. (2019) · Online Information Review · doi:10.1108/oir-10-2018-0332 · Cited by 64 · see citing works
Digital Phenotyping of Big Five Personality via Facebook Data Mining: A Meta-Analysis
Marengo, D., & Montag, C. (2020) · Digital Psychology · doi:10.24989/dp.v1i1.1823 · Cited by 64 · see citing works
The Autonomous Mind: The Right to Freedom of Thought in the Twenty-First Century
McCarthy‐Jones, S. (2019) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2019.00019 · Cited by 73 · see citing works
AI on the street: Context-dependent responses to artificial intelligence
Dorotić, M., Stagno, E., & Warlop, L. (2023) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2023.08.010 · Cited by 68 · see citing works
Theory In, Theory Out: The Uses of Social Theory in Machine Learning for Social Science
Radford, J., & Joseph, K. (2020) · Frontiers in Big Data · doi:10.3389/fdata.2020.00018 · Cited by 58 · see citing works
“For example, following misguided intuitions about sexuality and its causes can lead to incorrect claims, made on top of poor research design decisions, about the biological nature of sexuality (Wang and Kosinski, 2018).”
BackgroundThe compatibility of theoretical frameworks with machine learning analyses in psychological research
Elhai, J. D., & Montag, C. (2020) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2020.05.002 · Cited by 62 · see citing works
Bueno, C. C. (2019) · Theory Culture & Society · doi:10.1177/0263276419867752 · Cited by 58 · see citing works
In defence of machine learning: Debunking the myths of artificial intelligence
Laurent, C. D. S. (2018) · Europe’s Journal of Psychology · doi:10.5964/ejop.v14i4.1823 · Cited by 56 · see citing works
“When a team of data scientists at Stanford developed an algorithm capable of guessing, above chance, a person’s sexual preferences (Wang & Kosinski, 2017), they explained that their aim had only been to explore the potential dangers of AI (Levin, 2017).”
BackgroundAlgorithmic governmentality and the space of ethics: Examples from ‘People Analytics’
Weiskopf, R., & Hansen, H. K. (2022) · Human Relations · doi:10.1177/00187267221075346 · Cited by 67 · see citing works
“The idea of ‘revealing personality’, or inferring emotional states, or even sexual orientations (Kosinski and Wang, 2018) from measurable characteristics of the human face has often been criticized as ‘pseudo science’ in the tradition of phrenology and physiog-nomy (Metz, 2019).”
BackgroundThe Right to Erasure in EU Data Protection Law
Ausloos, J. (2020) · doi:10.1093/oso/9780198847977.001.0001 · Cited by 53 · see citing works
AI is Changing the World: For Better or for Worse?
Grewal, D., Guha, A., & Becker, M. (2024) · Journal of Macromarketing · doi:10.1177/02761467241254450 · Cited by 48 · see citing works
“For example, if an application can identify a person ’ s sexual orientation on the basis of their facial features, it could result in persecution by oppressive governments (Wang and Kosinski 2018).”
Automatic identification of myopia based on ocular appearance images using deep learning
Yang, Y., Li, R., Lin, D., et al. (2020) · Annals of Translational Medicine · doi:10.21037/atm.2019.12.39 · Cited by 54 · see citing works
“Facial recognition with deep convolution neural networks (DCNNs) has the potential to extract subtle features as well as hidden patterns, such as identifying facial phenotypes of genetic disorders and detecting sexual orientation (18,19).”
BackgroundGaydar and the Fallacy of Decontextualized Measurement
Gelman, A., Mattson, G., & Simpson, D. (2018) · Sociological Science · doi:10.15195/v5.a12 · Cited by 53 · see citing works
“As noted above, we are not claiming that experiments such as those of Tabak and Zayas (2012) and Wang and Kosinski (2018) are useless.”
MethodResultInfluentialTruth from the machine: artificial intelligence and the materialization of identity
Keyes, O., Hitzig, Z., & Blell, M. T. (2021) · Interdisciplinary Science Reviews · doi:10.1080/03080188.2020.1840224 · Cited by 57 · see citing works
“’ (Wang and Kosinski 2018, our emphasis).”
BackgroundInfluentialAI for Everyone? Critical Perspectives
Verdegem, P., Kaplan, A., Hofkirchner, W., Ng, J., McQuillan, D., Rehak, R., Daly, A., Devit, S. K., Mann, M., Steinhoff, J., et al. (2021) · University of Westminster Press eBooks · doi:10.16997/book55 · Cited by 55 · see citing works
Human Rights and Technology: New Challenges for Justice and Accountability
Land, M. K., & Aronson, J. D. (2020) · Annual Review of Law and Social Science · doi:10.1146/annurev-lawsocsci-060220-081955 · Cited by 56 · see citing works
Ban, H., & Ning, J. (2021) · Mobile Information Systems · doi:10.1155/2021/8670739 · Cited by 49 · see citing works
RetractedJemielniak, D. (2020) · doi:10.1093/oso/9780198839705.001.0001 · Cited by 63 · see citing works
Big Data, Biomedical Research, and Ethics Review: New Challenges for IRBs
Ferretti, A., Ienca, M., Hurst, S., & Vayena, E. (2020) · Ethics & Human Research · doi:10.1002/eahr.500065 · Cited by 46 · see citing works
Terhörst, P., Damer, N., Kirchbuchner, F., & Kuijper, A. (2019) · Applied Intelligence · doi:10.1007/s10489-019-01432-5 · Cited by 44 · see citing works
“Even the sexual orientation can be automatically obtained with an accuracy in the range of 83% to 91% [29].”
BackgroundPrivacy matters… or does It? Algorithms, rationalization, and the erosion of concern for privacy.
Fast, N. J., & Jago, A. S. (2020) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2019.07.011 · Cited by 42 · see citing works
Searching for gaydar: Blind spots in the study of sexual orientation perception
Miller, A. E. (2018) · Psychology and Sexuality · doi:10.1080/19419899.2018.1468353 · Cited by 44 · see citing works
“…intense in the fall of 2017 when two Stanford researchers published a study in the Journal of Personality and Social Psychology (JPSP) that concluded artificial intelligence (AI) software could accurately identify 81% of gay men and 71% of lesbian women by their faces (Wang & Kosinski, 2018).”
BackgroundBerger, S., & Baria, A. T. (2022) · Frontiers in Pain Research · doi:10.3389/fpain.2022.896276 · Cited by 42 · see citing works
“orientation (359), and criminality27, all of which are based on the debunked pseudoscience of physiognomy (360).”
BackgroundYaden, D., Eichstaedt, J., & Medaglia, J. (2018) · Frontiers in Psychology · doi:10.3389/fpsyg.2018.00962 · Cited by 41 · see citing works
Ethics and Good Practice in Computational Paralinguistics
Batliner, A. M., Hantke, S., & Schuller, B. W. (2020) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2020.3021015 · Cited by 42 · see citing works
“[63] claim that private traits, including sexual orientation, can be detected automatically via big data (Facebook, digitally available images, etc.”
Generating De-identification facial images based on the attention models and adversarial examples
Yang, J., Zhang, W., Liu, J., Wu, J., & Yang, J. (2022) · Alexandria Engineering Journal · doi:10.1016/j.aej.2022.02.007 · Cited by 43 · see citing works
Research Ethics, Vulnerability, and Trust on the Internet
Tiidenberg, K. (2019) · Springer eBooks · doi:10.1007/978-94-024-1555-1_55 · Cited by 43 · see citing works
Suppressing Gender and Age in Face Templates Using Incremental Variable Elimination
Terhörst, P., Damer, N., Kirchbuchner, F., & Kuijper, A. (2019) · International Conference on Biometrics · doi:10.1109/icb45273.2019.8987331 · Cited by 42 · see citing works
“Even the sexual orientation can be automatically obtained with an accuracy of around 83% to 91% [31].”
BackgroundPredicting individual-level income from Facebook profiles
Matz, S., Menges, J. I., Stillwell, D. J., & Schwartz, H. A. (2019) · PLoS ONE · doi:10.1371/journal.pone.0214369 · Cited by 41 · see citing works
Do, K., Pang, R. Y., Jiang, J., & Reinecke, K. (2023) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3581347 · Cited by 41 · see citing works
12th ACM Conference on Web Science
Shahsavari, S., Ebrahimzadeh, E., Shahbazi, B., Falahi, M., Holur, P., Bandari, R., Tangherlini, T. R., & Roychowdhury, V. (2020) · doi:10.1145/3394231 · Cited by 39 · see citing works
“…[16], who used deep neural networks to detect white males’ sexuality. controversial research implied that facial images of the LGBTQ population have distinct characteristics from the heterosexual population – LGBTQ population tends to have atypical gender expression and grooming styles [16].”
BackgroundImproving electron micrograph signal-to-noise with an atrous convolutional encoder-decoder
Ede, J. M., & Beanland, R. (2019) · Ultramicroscopy · doi:10.1016/j.ultramic.2019.03.017 · Cited by 38 · see citing works
Tenove, C., Buffie, J., McKay, S., & Moscrop, D. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3235819 · Cited by 42 · see citing works
Marres, N., Castelle, M., Gobbo, B., Poletti, C., & Tripp, J. (2024) · Big Data & Society · doi:10.1177/20539517241255103 · Cited by 37 · see citing works
- Authors
- Wang, Y., & Kosinski, M.
- Affiliations
- Stanford University (US)
- ORCID
- Kosinski, Michal
- Venue
- Journal of Personality and Social Psychology
- Published
- 2018-02
- Volume
- 114
- Issue
- 2
- Pages
- 246-257
- ISSN
- 0022-3514 (ISSN-L)
- 1939-1315
- OpenAlex
- W2781265533
- Cited by
- 766
- References
- 66
- Citations by year
- 2026: 29 · 2025: 57 · 2024: 57 · 2023: 146 · 2022: 98 · 2021: 92 · 2020: 155 · 2019: 75 · 2018: 51 · 2017: 4 · 2013: 1
- FWCI
- 55.54
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- cc-by
- Locations
- Journal of Personality and Social Psychology (publishedVersion)
- PubMed (publishedVersion)
- PsyArXiv (OSF Preprints) (submittedVersion)
- OSF Preprints (OSF Preprints) (submittedVersion)
- PsyArXiv (OSF Preprints) (submittedVersion, cc-by)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Evolutionary Psychology and Human Behavior
- Keywords
- face images, deep neural network, gay men, gay women, facial morphology, privacy, computer vision, logistic regression
- MeSH terms
- Facial Recognition, Deep Learning, Adult, Face, Female, Humans, Male, Social Perception, Sexuality
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
